An ontology-based Modelling framework for detailed spatio-temporal population estimation

Abstract

The whereabouts of population changes over short time scales as people go about their daily lives. A requirement for very detailed population estimates that reflects this variation has been recognised for decades, with myriad application areas that could benefit from this in the public, research and commercial domains. Yet there remains a lack of suitable, extensible and transferrable methods for estimating population at the fine spatial and temporal scales of detail required for these applications. Such population estimation requires the integration of data from diverse sources including core geographic, statistical and the new and emerging sources from sensors and the internet. This integration includes creating appropriate linkages between the spatial, temporal and attribute data domains where these are related. Semantic web technologies provide a simple data model for the integration of such diverse data. Ontologies provide the ability to formalise the relationships between these data and make inferences through those defined relationships. This thesis presents a framework, or structure, into which new, evolving and alternative data can be worked with the goal of generating population estimates at very fine spatial (address level) and temporal (continuous) detail. The three-part modelling framework presented here integrates population in the spatial, temporal and attribute domains to estimate population counts at the level of addresses, on a continuous temporal scale. This thesis introduces, for the first time, the foundations of a semantic web-based modelling solution to this problem in the population estimation domain.

More information

Identifiers

ORCID for Rebecca King: orcid.org/0000-0003-3953-3413

ORCID for Nicholas Gibbins: orcid.org/0000-0002-6140-9956

ORCID for David Martin: orcid.org/0000-0003-0397-0769

Catalogue record

Export record